Graph representations for biology and medicine
EE-626
Media
This file is part of the content downloaded from Graph representations for biology and medicine .
Summary: Systems of interacting entities, modeled as graphs, are pervasive in biology and medicine. The class will cover advanced topics in signal processing and machine learning on graphs and networks, and will showcase applications of the tools in biomedicine. It will be held as an advanced seminar, which will familiarize students with recent developments in the topic, through a combination of lectures on some fundamentals on processing and analyzing data on graphs, and the presentation of original research articles that make use of these tools for scientific advances in biology and medicine.
When: Every Wednesday 13:15-15:00
Where: INF 019
Week 1: Graph representations for biology and medicine - Introduction
Background material:
- Li et al., Graph representation learning in biomedicine and healthcare, Nature Biomed. Engineering, 2022
- Zitnik et al., Current and future directions in network biology, arXiv, 2023
- Bassett et al., Network neuroscience, Nature neuroscience, 2017
- A. Avena-Koenigsberger et al., Communication dynamics in complex brain networks, Nature Rev Neurosci, 2018
- Barabasi et al., Network medicine: a network-based approach to human disease, Nature Rev Genetics, 2010
- Johnson et al., Graph Artificial Intelligence in Medicine, Annual review of biomedical data science, 2024
- Altucci et al.,Artificial Intelligence and Network Medicine: Path to Precision Medicine, NEJM AI, 2025
Week 2: Quick introduction into graph machine learning (Part A)
Week 3: Quick introduction into graph machine learning (Part B)
We will continue with the slides from last week.
Week 4: Graph generative models
Week 5: Graph generative models: Examples of architectures
We will continue the slides from last week.
Week 6: Spatially resolved (multi)omics
The following papers will be discussed:
- Ali et al., Graph neural networks learn emergent tissue properties for spatial molecular profiles, Nature Communications, 2025
- Bao et al., Spatially informed graph transformers for spatially resolved transcriptomics, Nature Communications Biology, 2025
Presenter: Theo
Discussion leaders: Amaury, Yves
Week 7: Histopathology and tumor microenvironment
The following papers will be discussed:
- Shao et al., Tumor micro-environment interactions guided graph learning for survival analysis of human cancers from whole-slide pathological images, CVPR, 2024
- Xu et al., TopoCellGen: Generating histopathology cell topology with a diffusion model, CVPR, 2024
Presenters: Andrea, Yves
Discussion leaders: Vasiliki, Theo
Week 8: Neuroscience
The following papers will be discussed:
- Qiu et al., Towards Graph Neural Networks with Domain-Generalizable Explainability for fMRI-Based Brain Disorder Diagnosis, MICCAI, 2024
- Afzal et al., REST: Efficient and Accelerated EEG Seizure Analysis through Residual State Updates, ICML, 2024
Presenter: Alejandro
Discussion leads: Tian, Adrian
Week 9: [Invited lecture - ELE 117] From brains to hearts, graph-based machine learning as a unifier for prediction on spatio-temporal medical imaging data
Speaker: Dr. Jonas Richiardi, UNIL, CHUV
Abstract: Interactions between parts of a biological system are a central concept in biomedical sciences. These interaction networks can be mathematically modelled as labelled graphs, and this type of approach has been used across scales for genes, proteins, cells, organs, or individuals. Coupled with machine learning, graph representations have many promising applications for precision medicine, including differential diagnosis, treatment planning, survival modelling, and prognosis.
In this talk, we will discuss how graph-based learning can be developed and applied to spatio-temporal medical imaging data, with a focus on brain and heart imaging. We will start with an introduction to spatio-temporal imaging data, then show how "organ graphs" can be defined and computed for both brain and heart, yielding an expressive and compact meso-scale representation of each patient. We'll introduce a new statistical estimator for spatio-temporal correlation in brain imaging, the local correlation of averages, which exhibits superior theoretical and empirical properties.
With representations addressed, we will then transition to defining the machine learning tasks that can be addressed given an organ graph representation, and give a rapid overview of existing approaches. As an example of a novel approach, we'll focus on a newly proposed graph neural process model, which combines multiplex graphs with neural ordinary differential equations and neural processes to yield promising performance in spatio-temporal trajectory reconstruction, interpolation, and graph classification for cardiac imaging data, as well as a latent space that reflects known underlying pathophysiological features of cardiac disease.
Finally, we will show some recent empirical benchmark results on graph neural networks for regression on brain graphs, with applications to multimodal graphs, (where edges are vector-valued), transfer learning between regression tasks, and hyperbolic graph neural networks.
Short bio: Jonas Richiardi is a Principal Investigator and Senior Lecturer at the Department of Radiology, Lausanne University Hospital, Switzerland, and heads the Translational Machine Learning Laboratory (https://unil.ch/tml). He is also the section head of the Imaging for Precision Medicine section (https://cibm.ch/research/projects/imaging-for-precision-medicine/), part of the Data Science Module of the CIBM Center for Biomedical Imaging.
Previously, he was Clinical Research Lead at Siemens Healthcare, a Marie Curie fellow in Neurology at Stanford University and the University of Geneva, and a post-doctoral researcher in the Medical Image Processing Lab (EPFL/UNIGE). He obtained his Ph.D. at EPFL in the Laboratory of the Dalle Molle Institute for Perceptual Artificial Intelligence, Signal Processing Institute, and and his M.Phil. from the university of Cambridge's Engineering Department and Computer Laboratory.
His research interests include machine learning for complex multimodal biological data, in particular magnetic resonance brain imaging data and its combination with -omics data. Methods development are focused on graph-based machine learning approaches for spatio-temporal imaging data, learning from scarce and heterogeneous data, and multimodal approaches. Applications to precision medicine include diagnosis, treatment selection, prognosis, and treatment response prediction, in particular for stroke, cardiovascular disease, and oncology. In parallel, he leads efforts to develop imaging data science infrastructure so that these techniques can be applied to messy, hospital-scale clinical routine data.
Location: Please note that the talk will take place in ELE 117
Abstract: Interactions between parts of a biological system are a central concept in biomedical sciences. These interaction networks can be mathematically modelled as labelled graphs, and this type of approach has been used across scales for genes, proteins, cells, organs, or individuals. Coupled with machine learning, graph representations have many promising applications for precision medicine, including differential diagnosis, treatment planning, survival modelling, and prognosis.
In this talk, we will discuss how graph-based learning can be developed and applied to spatio-temporal medical imaging data, with a focus on brain and heart imaging. We will start with an introduction to spatio-temporal imaging data, then show how "organ graphs" can be defined and computed for both brain and heart, yielding an expressive and compact meso-scale representation of each patient. We'll introduce a new statistical estimator for spatio-temporal correlation in brain imaging, the local correlation of averages, which exhibits superior theoretical and empirical properties.
With representations addressed, we will then transition to defining the machine learning tasks that can be addressed given an organ graph representation, and give a rapid overview of existing approaches. As an example of a novel approach, we'll focus on a newly proposed graph neural process model, which combines multiplex graphs with neural ordinary differential equations and neural processes to yield promising performance in spatio-temporal trajectory reconstruction, interpolation, and graph classification for cardiac imaging data, as well as a latent space that reflects known underlying pathophysiological features of cardiac disease.
Finally, we will show some recent empirical benchmark results on graph neural networks for regression on brain graphs, with applications to multimodal graphs, (where edges are vector-valued), transfer learning between regression tasks, and hyperbolic graph neural networks.
Short bio: Jonas Richiardi is a Principal Investigator and Senior Lecturer at the Department of Radiology, Lausanne University Hospital, Switzerland, and heads the Translational Machine Learning Laboratory (https://unil.ch/tml). He is also the section head of the Imaging for Precision Medicine section (https://cibm.ch/research/projects/imaging-for-precision-medicine/), part of the Data Science Module of the CIBM Center for Biomedical Imaging.
Previously, he was Clinical Research Lead at Siemens Healthcare, a Marie Curie fellow in Neurology at Stanford University and the University of Geneva, and a post-doctoral researcher in the Medical Image Processing Lab (EPFL/UNIGE). He obtained his Ph.D. at EPFL in the Laboratory of the Dalle Molle Institute for Perceptual Artificial Intelligence, Signal Processing Institute, and and his M.Phil. from the university of Cambridge's Engineering Department and Computer Laboratory.
His research interests include machine learning for complex multimodal biological data, in particular magnetic resonance brain imaging data and its combination with -omics data. Methods development are focused on graph-based machine learning approaches for spatio-temporal imaging data, learning from scarce and heterogeneous data, and multimodal approaches. Applications to precision medicine include diagnosis, treatment selection, prognosis, and treatment response prediction, in particular for stroke, cardiovascular disease, and oncology. In parallel, he leads efforts to develop imaging data science infrastructure so that these techniques can be applied to messy, hospital-scale clinical routine data.
Location: Please note that the talk will take place in ELE 117
Week 10: Medical imaging
The following papers will be discussed:
- Khalid Faizi et al., Graph neural network model using radiomics for lung CT image segmentation, Nature Scientific Reports, 2025
Ding et al., Combining graph neural network and Mamba to capture local and global tissue spatial relationships in whole slide images, Nature Scientific Reports, 2025
Presenters: Amaury, Kevin
Discussion leads: Andrea, Alejandro
Week 11: Context-aware learning
The following papers will be discussed:
- Li et al., Contextual AI models for single-cell protein biology, Nature Methods, 2024
- Zhang et al., EquiPocket: an E(3)-Equivariant Geometric Graph Neural Network for Ligand Binding Site Prediction, ICML 2024
Presenter: Jiying
Discussion leads: Andrea, Yves
Week 12: Learning from multi-modal/multi-view data
The following papers will be discussed:
- Wu et al., Leveraging Tumor Heterogeneity: Heterogeneous Graph Representation Learning for Cancer Survival Prediction in Whole Slide Images,Leveraging Tumor Heterogeneity: Heterogeneous Graph Representation Learning for Cancer Survival Prediction in Whole Slide Images, NeurIPS, 2024
- Yang et al., STAIG: Spatial transcriptomics analysis via image-aided graph contrastive learning for domain exploration and alignment-free integration, Nature Communications, 2025
Presenter: Vasiliki
Discussion leaders: Alina, Tian
Week 13: Drug discovery (topic postponed to next week)
The following papers will be discussed:
- Fang et al, ATOMICA: Learning Universal Representations of Intermolecular Interactions, bioarxiv, 2025
- Wu et al., Surface-based molecular design with multi-modal flow matching, KDD, 2025
Presenters: Adrian, Tian
Discussion leaders: Amaury, Kevin
Week 14: Drug discovery & Learning from EHR data
The following papers will be discussed:
- [Drug discovery] Wu et al., Surface-based molecular design with multi-modal flow matching, KDD, 2025
- [Drug discovery] Fang et al., ATOMICA: Learning Universal Representations of Intermolecular interactions, bioArxiv, 2025 [online video - see below]
- [EHR] Poulain et al., Graph Transformers on EHRs: Better Representation Improves Downstream Performance, ICLR, 2024
The video presentation for ATOMICA can be downloaded from here.
Presenters: Tian, Adrian, Alina
Discussion leads: Amaury, Kevin, Jiying